[unreadable] The long term goal of the applicant is to develop new, validated biomarkers that will enable early detection of vascular disease. More than 65 million people in the United States are affected by hypertension. Arteriosclerosis (i.e., atherosclerosis and arteriolosclerosis) of the cerebral, cardiac, renal, and peripheral arterial beds is accelerated in the setting of hypertension and leads to target organ damage and [unreadable] cardiovascular sequelae such as stroke, myocardial infarction, chronic renal insufficiency, and claudication. Currently available risk assessment tools have poor sensitivity and specificity. Validating new biomarkers of arteriosclerotic vascular disease will allow formulation of innovative preventive and therapeutic strategies to ultimately improve quality-of-life and reduce mortality. This application proposes to investigate the [unreadable] relationship of 50 candidate proteomic markers in pathways of inflammation, thrombosis, lipoprotein metabolism, vascular remodeling, and oxidative stress to 4 different quantitative measures of arteriosclerosis in African American (n = 1150) and non-Hispanic white (n = 1150) individuals belonging to hypertensive sibships, leveraging the excellent clinical research infrastructure developed as part of the NHLBI's Family Blood Pressure Program. The measures of atherosclerotic or 'macrovascular' disease available for this proposal include the ankle-brachial index and quantity of coronary artery calcification. The measures of arteriolosclerotic or 'microvascular' disease available for this proposal include cerebral leukoaraiosis assessed by magnetic resonance imaging of the brain and urinary albumin-creatinine ratio. We will accomplish the following specific aims: Specific Aim 1: We will develop robust validated assays for 50 candidate proteomic markers and using the validated assays, measure the proteomic markers in 2300 subjects belonging to the two ethnic cohorts. We will also develop and validate panels of 6-8 markers for multiplex assays. Specific Aim 2: We will use a variety of statistical approaches including Principal Components Factor Analysis and (Bayesian and non-Bayesian) model-averaging to identify the best predictive models (from among the 50 proteomic markers) for each of the 4 quantitative measures of arteriosclerosis in the 2300 subjects belonging to the two ethnic cohorts. [unreadable] [unreadable]